Children of farmworkers in California are exposed to a unique mix of chemicals, including polybrominated diphenyl ether (PBDE) flame retardants, which are prevalent in California due to the state's strict flammability standards; manganese (Mn), which is a key component of the widely used agricultural fungicides maneb and mancozeb, and the pesticide dichlorodiphenyltrichloroethane (DDT), which was recently used in Mexico where most of their farmworker parents originated. In this proposal, we continue our ongoing research into developing state-of-the-art methods of assessing children's environmental exposures to neurotoxic and endocrine-disrupting compounds and characterizing children's levels, sources, and routes of exposure. We propose to use shed deciduous teeth to develop and validate new biomarkers of prenatal and early postnatal exposure to Mn and will explore whether PBDEs and DDT/E can also be measured in teeth. Using our extensive biorepository of banked prenatal and early postnatal samples, we will determine the relationship between Mn, PBDE, and DDT/E levels in teeth with levels in blood, urine, and breastmilk. We will quantify the sources of children's Mn exposure by measuring levels of Mn in 450 house dust samples collected during the prenatal period and identifying factors that explain these levels, including farmwork of household members, nearby agricultural maneb and mancozeb use, traffic intensity, industrial emissions, and wind patterns. We will examine these factors, as well as diet, in relation to prenatal and early postnatal Mn levels measured in teeth from CHAMACOS children (n=300) and childhood Mn levels measured in blood from CHAMACOS boys collected at age 9 (n=300). We will use innovative GIS methods, remote sensing technology, and Bayesian modeling to build statistical and predictive models of Mn exposure. Finally, we propose to identify population-level correlates of PBDE and DDT/E exposure measured in blood collected from 9-year-old boys (n=300). Factors of interest include diet, breastfeeding history, consumer electronic use, SES, child BMI, time outside the US, household characteristics, and housing quality. This proposal utilizes a rich dataset and biorepository of information on Mexican-American children participating in the CHAMACOS longitudinal birth cohort in the Salinas Valley, California. Novel methods developed under this proposal have the potential to revolutionize our ability to assess prenatal and early postnatal exposures in epidemiologic studies of children's environmental health.